DC4 Co-simulation strategies involving IBRA for solution of multi-field problems:

Complex technical systems often require a partitioned approach to enable disciplinary modelling and simulation with bestsuited solution approaches and discretization techniques in each domain.

Developing Flexible and Accurate Strategies for Multiphysics Partitioned Simulations

As part of the GECKO Project, my PhD research focuses on developing computational methods for partitioned multi-disciplinary simulations, with particular attention to Fluid-Structure Interaction (FSI) problems. The aim is to address the challenges of coupling different discretization techniques while ensuring robust, accurate, and scalable simulations that can be applied to a variety of engineering problems.

One of the key objectives of this work is to enhance the efficiency and flexibility of partitioned simulation approaches. These methods allow different physical domains, such as fluids and structures, to be modeled using specialized solvers, which are then coupled and exchange data across their interfaces. This requires the development of advanced data transfer operators that ensure accurate mapping between domains, even when the underlying discretizations differ significantly.

During the first year, significant progress was made in tackling these challenges. Among the main achievements is the implementation of robust coupling schemes that integrate Isogeometric B-Rep Analysis (IBRA) with other solvers, leveraging IBRA’s ability to provide highly accurate geometric representations. Additionally, efforts were focused on improving data transfer methods, particularly through the implementation of the Mortar Mapper, which is capable of handling both body-fitted and unfitted discretizations.

Another important outcome of this research is a new approach to addressing singularity issues that arise when imposing strong boundary conditions on unfitted meshes. This development expands the applicability of partitioned simulations, enabling accurate and reliable modeling in scenarios involving unfitted discretizations. The methods developed during this first phase have been implemented within the Kratos-Multiphysics open-source framework, ensuring accessibility and further development opportunities.

The results achieved so far provide a strong foundation for addressing more complex engineering problems in fields such as aerospace, biomechanics, and civil engineering. Future work will involve publishing these methods, comparing the performance of different data transfer operators, and applying the techniques to benchmark cases to validate their efficiency and accuracy further.

The research represents a step forward in providing tools that enhance the reliability and flexibility of simulations for multi-physics problems, contributing to the design of safer and more efficient engineering solutions.